Ga direct naar de content

Thomas Philippon: “For fintechs, the biggest cost of entry is legal uncertainty”

Geplaatst als type:
Gepubliceerd om: september 21 2017

Thomas Philippon is a Professor of Finance at New York University Stern School of Business. The IMF named him as one of the 25 best economists under the age of 45, and he received the Bernácer Prize for the best European economist under the age of forty in macroeconomics and finance. He has published over twenty papers, mainly on costs and efficiency in the financial sector, and is an advisor to the Federal Reserve Bank of New York on Monetary Policy.

He passionately believes that improvements in information technology do not benefit the customers of financial services as much as they should – as, for more than a century, the unit cost of financial intermediation has been between 1.5 to 2.0 percent (Philippon, 2015) – and that financial services could be a lot cheaper. “In that respect, the puzzle is not that fintech is happening now. The puzzle is why it did not happen earlier.” (Philippon, 2016)

“Payment services are finance’s fundamental building block. In essence, every action in finance boils down to two things: information and pipes to carry the information. So, everything you do in finance is a mixture of creating, managing and interpreting information and sending it through the pipes. The pipes of the payment systems are among the most important ones.

The banks seem less scared of fintechs today than a few years ago

It’s important to note that payment services that exist between individuals are completely different from those existing between large players. If you say payments services, you’re not only referring to million-dollar bank-to-bank transactions but also to people paying for a cookie with their iPhone.”

How important is innovation in payment services?

“In payment systems, fintech is now ubiquitous for three reasons. First, availability of capital is less of an issue. It’s easy to get the startup capital in advance and you can in fact rent anything you need. Client acquisition is key.

Second, payment services provide a natural entry point for many fintechs, because payments are a market with lots of small transactions. For a new player this is ideal, as it provides customers with opportunities to learn about new players.

Take the difference between payment services and money management. If you have a lot of money in your savings account, you’re not going to take the risk of trying a new player, because if it fails you will have lost a lot of money. But if you’ve got an app on your phone and the worst-case scenario is that you may to lose five dollars, you would probably be a lot more willing to experiment. Especially the younger generation.

And third, especially cross-border payments used to be outrageously expensive and inefficient. People lost ten to twenty percent of what they earned on transaction costs. Fintechs in payment services can decrease those costs.”

Thomas PhilipponHH/Leemage

Unfortunately, compared with fields like banking, there isn’t a lot of empirical research on fintech.

“First, that’s a sign that previously the industry was so closed before.

Second, academic research in finance is too focused on to hedge-fund style research: trying to find alfa strategies in bigdata. Hedge funds would do that research themselves and we shouldn’t pay people in academia to do that. And there is not enough research asking questions about efficiency and welfare in the financial system. When I started my research on the efficiency of finance, I was the first to develop a measure in order to get a rough sense of how the efficiency of financial intermediation is moving over time. That should have been done years ago.”

Do you think fintech startups are going to lead to the unbundling of banks?

“The unbundling part is not as obvious as it sounds. There’s going to be some unbundling, but banks are very likely to focus on their core activities and on where they add value. The rest they can outsource.

The banks seem less scared of fintechs today than a few years ago, because they realize they’re actually able to compete. All they needed to do was offer products that people liked and all the big banks now have their own online entities.

A good example is the customer interface, which is not the core of the banking business. If a startup has a great idea for improving the website or the interaction between the bank and its customers, they’ll sell the product to all of the banks. But for banks that’s merely the tip of the iceberg. And, as to that particular part of their process, they’ll buy the innovation from the fintech in order to improve user experience. The data part of banking is going to be more important and the cross-subsidizing between the different products they sell is going to change. But I think banks have more than enough firepower to fight back.”

There is some concern that after the current market phase, with all these new entrants in payment services, there will be a move towards concentration. Do you believe there’s going to be enough competition?

“On a national level there may be a natural monopoly, but on a European level I’m not so worried about concentration. On the European market, there are going to be many players able to provide payment services. One might imagine various types of players providing payment services, and as long as a few of them are competing you’ll be fine.

As a regulator, you need a rough idea of the players that are going to be active on the European market

As a regulator, you need a rough idea of who these players might be. Payments systems are like messaging, and if you’re good at doing messaging you should be good at providing payments systems. Therefore, I expect that several social networks, several telecom operators and maybe some other players might move into payment services if they wanted to. If Facebook wanted to be a bank, they would have an amazing headstart in many dimensions. The data they have are mindblowing and, with onboarding [integrating new employees into their organization, ed.] and things like Artificial Intelligence, they are far better than any other organization. They just don’t want to become a bank because of the regulation side. Furthermore, the other companies from the US – Google, Apple and Amazon – are going to be around and will provide payment services. And, from China, Alibaba and Tencent can also become active on the European market. They already deal with billions of consumers. The other players in Europe who might do pretty well are in the telecom industry. Because they have the data, the capital and the tech guys, and they are pretty good at doing these things. So that’s a lot of companies potentially competing.

The strategic decision for Europe is: do we want at least some of these players to be European? That’s a fair discussion to have, because maybe you don’t want lose control of your payment system. Payments are the pipes, and the ECB must be able to trust the pipes and to control them.”

Does that mean regulations should be made at a European Union level?

“For anything that revolves around competition, regulating at EU level is a better idea, and then you can delegate the nitty-gritty details (local specificities) to the national level. The recent development in Europe as to the access to your own data shows that, at EU level, it really works. They’ve decided that any information in your bank account doesn’t belong to your bank, but belongs to you. That’s a high level principle, so that’s an EU job. It wouldn’t have had the same traction if it were done at a national level. In the US for example, all the banks are against it. So, it’s pretty clear who is doing the better job here.”

Still, regulators are uncertain about how the market is going to develop, yet they need to implement regulation now, for instance in order to encourage competition. Isn’t there also a risk that you might unintentionally take a decision that will prevent the market from developing?

“That’s the classic trade-off between innovation and regulation. Given the track record of finance, there are two things that are critical.

One thing to keep in mind is that the issues in finance and in every crisis are always the same, in the sense that there is leverage hidden somewhere, which is picked up by big players and causes a liquidity risk. You want to make sure that, in whatever system you have, there is lower leverage, or that you’re able to control the leverage.

For me, the money market funds (MMF) are best example of where there should have been regulation early on. The MMFs were badly regulated from the start, because they were allowed to operate like banks, only without banking regulation. It was bad idea not to regulate them early on, because it’s much easier to regulate something that is small.

The other important thing is that, nowadays, the biggest cost of entry is legal uncertainty. Entrants don’t know what their liabilities are, and the regulators can never give them a straight answer. Take for example the people who wanted to make a website comparing operating fees for asset management. The website would ask a client about their current portfolio and then would show them a list of portfolios with the exactly the same risk exposure within a certain statistical distance, ranking them as to annual fees. In terms of finance, that’s the same product. They could never get it off the ground though, because there was never a regulator who would say the firm wasn’t liable for the choices people made on the basis of this information.”

And should the regulator’s focus then be on individual entities or on function?

“We should all agree that, if you think about it from first principle, it’s function that matters. And therefore markets should be regulated on the basis of function. But at some point pragmatism comes into it, and in practice many of these regulations are going to be on the firm level, simply because the large firms are also the ones posing a systemic risk.

But there is another side to the debate, which I think is more important. That’s the level playing field. When people say we should regulate by function, what they really have in mind is that, if two entities do the same thing, they should have the same regulations. However, here I strongly disagree, because that’s a very narrow view of level playing field. There’s no level playing field if you have the same regulation for a gigantic bank, providing a large number of financial services, versus a tiny startup, providing only one of those services. If you apply the same regulation to both, you’re going to kill the startup. So, the rules have to be proportional and have to take into account that some firms are big, have years of experience in dealing with regulations and have too-big-to-fail guarantees.

The reason why financial intermediation is so expensive is that there isn’t enough entry, because the rules were never designed to encourage entry. If anything, they were designed to prevent it. When there’s entry, like there is in online banking, the reason new players can exist is because the regulators have accepted to make the rules easier for them.”


Philippon, T. (2015) Has the finance industry become less efficient? American ­Economic Review, 105(4), 1408–1438.

Philippon, T. (2016) The FinTech opportunity. NBER Working Paper, 22476.